The incidence of epilepsy rises rapidly after age 60 with a reported prevalence of 1.48% for persons over 75. Pharmacoepidemiologic studies performed during the current P50 show that the prevalence of anti epileptic drug (AED) use is approximately 2% among the community-dwelling elderly (Health Care Finance Administration data set (n = ~12,000/yr) and approximately 9% among nursing home residents (Beverly data set). Clearly, the absolute number of elderly using AEDs will increase greatly in the near future, but very little is known about the safe and effective use of these powerful modulators of neurological functioning in this population. The goal of this 5-year program project is to increase our knowledge of the pharmacokinetics (PK), pharmacodynamics (PD), and pharmacoepidemiology (PE) of AEDs in the elderly. This information is critical for the design of appropriate controlled trials and the development of evidence-based guidelines for this population. We will investigate these issues using a variety of approaches including hepatic microsomes in vitro liver, stable labeled isotopes in volunteers, non-linear mixed effect modeling (NOME) and application of estimated AED concentrations to outcomes in a large nursing home population. Extensive genotyping of CYP3A4/5 and initial investigation of genetic and racial factors in AED metabolism will be carried out. The PK characteristics of carbamazepine (CBZ), valproic acid (VPA), lamotrigine (LTG), and topiramate (TPM) in elderly patients will be determined using stable labeled intravenous preparations developed by our group. NONMEM analysis will be performed using clinical data from Minnesota, Atlanta, Miami, and the national Veterans Administration Cooperative Study. These 3 sites will provide the largest group of elderly persons with epilepsy ever studied and have enough power to test the hypotheses. Findings regarding phenytoin (PHT), CBZ and VPA PK from the current and proposed P50 will be applied to dose, weight, and to outcome measures in the extant Beverly Nursing Home database. This will allow us to estimate AED concentrations in this population and determine prevalence and type of adverse events associated with high, middle, or low concentrations. The information obtained from this multi-pronged effort will have profound impact on the use of AEDs in the elderly.